About
This report is intended to act as a first-pass exploratory analysis
of the variants in your dataset. Looking for points that are
e.g. extreme outliers in the PCA plot and the depth/missingness plot are
a good way to identify potentially problematic (or interesting!)
samples. Several basic analysis are included, many of which are colored
by the result of a k-means clustering on the first two principal
components. A map is possible to generate if you place a tab-separated
file with sample ID, longitude, and latitude (decimal degrees) with the
extension .coords in the directory the contains all of the
QC output. By default in this pipeline, these are in the subdirectory
06_QC. By default, three populations are assumed for
k-means clustering, which may be a sensible starting point for visual
inspection of your results, but this value can be changed in the config
file.
